When a server goes down, it’s rarely just one thing — a RAID that dropped below its redundancy, a failed controller, a botched rebuild, or a physical fault in the drives themselves. Because a server is usually the heart of a business, that’s an emergency, and it’s exactly the kind of work we do most. We recover Dell, HP and Lenovo servers and their arrays — physical, RAID and virtualised — reassembling everything off imaged disks. Whether you searched server data recovery or RAID server recovery, the diagnosis is free and no-recovery-no-fee applies.
A forced rebuild onto a degraded server array is the fastest way to lose what was recoverable. Power it down, keep the drive-bay order, and let us image before anything is written back.
Most server disasters are a second failure landing on a first that went unnoticed — a RAID 5 quietly loses a disk, keeps running degraded, and weeks later a second disk drops and the array goes offline. Others are the controller: a failed RAID card or backplane scrambles the array metadata, or a rebuild started with disks in the wrong order overwrites the parity. And a healthy array can still be stranded by a dead motherboard or power fault. None of these is fixed by the server’s own rebuild — that step is usually what turns a recoverable outage into a permanent one.
The server’s array is never worked live. Every member disk is imaged individually onto our own storage — including any disk the controller has flagged as failed, because it often still holds the parity that makes the rest readable. From those images we reconstruct the array in software, working out the disk order, stripe size and parity by hand, then extract the volumes, databases and virtual machines from the rebuilt set. Because everything is done from copies, a mistake in reconstruction costs nothing — and the original disks leave exactly as they arrived.
We recover Dell PowerEdge, HPE ProLiant, Lenovo ThinkSystem, Supermicro and self-built servers, on Dell PERC, HP Smart Array, LSI/Broadcom and Adaptec controllers, running Windows Server, Linux, VMware ESXi or Hyper-V. Keep the drive-bay order intact, and send the whole machine where you can — the controller and its configuration matter as much as the disks.
Usually. A single disk failure shouldn’t take an array down, so a total outage normally means a second disk failed or the controller has faulted. We image every disk — including the failed ones — and rebuild the array off the hardware, which recovers the large majority of these.
Yes — a dead controller leaves healthy disks stranded, and we rebuild the array from the disks themselves rather than depending on the controller. We just need the disks in their original bay order, and ideally the server so we can match the configuration.
No. A rebuild onto a degraded array, or one started with the disks in the wrong order, is the single most common cause of permanent server data loss. Power it down and let us image every disk first.
Yes — once the underlying array is rebuilt, we extract the VMs, SQL or Exchange databases and file shares from it. If a specific database or VM is corrupt on top of a healthy array, we handle that as part of the same job.
A free diagnosis first, always — then a fixed written quote before any work begins, and no fee at all on most jobs if the data doesn’t come back. Server and array recoveries are quoted after diagnosis, since the price depends on the number of disks and the fault — but you’ll always have a fixed figure in writing before any chargeable work. Drop your device at our Leeds address in The Pinnacle, or post it in fully insured from anywhere in Yorkshire; whichever route it takes, it’s handled in-house by our own engineers and never outsourced.
The diagnosis is free, the quote is fixed in writing, and most jobs carry no fee unless your data comes back. Stop any rebuild, keep the bay order, and get the disks to us — that protects what’s recoverable.